DocumentCode :
2995270
Title :
Fast nonlinear adaptive filtering using a partial window conjugate gradient algorithm
Author :
Birkett, A. Neil ; Goubran, Rafik A.
Author_Institution :
Dept. of Syst. & Comput. Eng., Carleton Univ., Ottawa, Ont., Canada
Volume :
6
fYear :
1996
fDate :
7-10 May 1996
Firstpage :
3541
Abstract :
In this paper a modified form of the partial conjugate gradient algorithm is presented for use in nonlinear filtering using neural networks. The algorithm is based on using a gradient average window to provide a trade-off between convergence rate and complexity which, depending on the choice of averaging window, is (in both complexity and speed of convergence) intermediate between the conventional backpropagation (BP) algorithm and the Newton methods. An additional simplification is introduced by replacing the calculated optimum step size αk by a normalized step size α¯, in the same manner as the normalized LMS algorithm. This new algorithm is applied to a cascaded neural network/nonlinear least mean squares structure for the identification of a nonlinear system. This proposed algorithm demonstrates improved convergence rates with even small choices of window size
Keywords :
adaptive filters; cascade networks; computational complexity; conjugate gradient methods; convergence of numerical methods; digital filters; feedforward neural nets; least mean squares methods; multilayer perceptrons; nonlinear filters; cascaded neural network; complexity; convergence rate; fast nonlinear adaptive filtering; gradient average window; identification; neural networks; nonlinear least mean squares; nonlinear system; normalized step size; partial window conjugate gradient algorithm; Adaptive filters; Backpropagation algorithms; Computer networks; Convergence; Cost function; Error correction; Filtering algorithms; Neural networks; Newton method; Nonlinear systems;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1996. ICASSP-96. Conference Proceedings., 1996 IEEE International Conference on
Conference_Location :
Atlanta, GA
ISSN :
1520-6149
Print_ISBN :
0-7803-3192-3
Type :
conf
DOI :
10.1109/ICASSP.1996.550793
Filename :
550793
Link To Document :
بازگشت